AC Optimal Power Flow Load Profiles and Solutions

Comprising load profiles and their respective optimal primal and dual solutions generated for AC Optimal Power Flow (OPF) research. It was created using the OPFLearn.jl Julia package, which employs a relaxed AC OPF formulation to produce datasets more representative of the feasible load space. The dataset is provided by the Department of Energy.

Use Cases

• Train machine learning models to predict optimal primal solutions from load profiles for power system optimization.
• Analyze the relationship between load profiles and dual solutions to study pricing and constraint marginal costs.
• Use the standardized load profiles for benchmarking AC OPF algorithms and solution methods.
• Develop surrogate models to approximate the nonlinear AC OPF problem using the provided primal and dual variable data.

Strengths

• Dataset creation method is documented in a peer-reviewed publication and an open-source package.
• Designed to be more representative of the AC OPF feasible load space compared to typical methods.
• Includes both optimal primal and dual solutions for each load profile, supporting a range of analytical tasks.

Limitations

• The specific number of rows, columns, and data size is unknown, limiting assessment of scale.
• Sample data is unavailable, preventing inspection of data structure and quality before download.
• The dataset's representativeness is tied to the specific relaxed AC OPF formulation used in its generation.

Provenance

Source

Department of Energy

Collection Method

Generated using the OPFLearn.jl Julia package, which processes load samples with AC OPF formulations from PowerModels.jl.

Freshness

Last updated on 2026-03-21.